1. Patent Search
  2. Details

KIT FOR PREDICTING OR DIAGNOSING NONALCOHOLIC FATTY LIVER DISEASE, AND METHOD FOR DIAGNOSING NONALCOHOLIC FATTY LIVER DISEASE

20220276246 · 2022-09-01

    Inventors

    Cpc classification

    International classification

    Abstract

    The present invention relates to a kit for predicting or diagnosing the degree of risk of disease, a method for providing information for predicting or diagnosing the degree of risk of disease, a method for screening a therapeutic agent of disease, and a pharmaceutical composition for prevention or treatment of disease, for nonalcoholic fatty liver disease. Specifically, through the kit for predicting or diagnosing of the present invention, the degree of risk of nonalcoholic fatty liver disease can be effectively predicted or diagnosed, and in particular, the predictive value and significance of information provision in non-obese subjects are excellent. Therefore, by providing effective information on nonalcoholic fatty liver disease through this, it can be effectively used to prevent or treat the corresponding disease. In addition, the pharmaceutical composition for prevention or treatment of the present invention may be effectively used for treatment of nonalcoholic fatty liver disease.

    Claims

    1. A kit for predicting or diagnosing a degree of risk of nonalcoholic fatty liver disease comprising a detection means detecting one or more detection markers selected from the group consisting of (a) one or more detection markers selected from the group consisting of microbial biomarkers of nonalcoholic fatty liver disease; (b) one or more detection markers selected from the group consisting of total bile acid and components of bile acid; and (c) intestinal short chain fatty acid.

    2. The kit according to claim 1, wherein the (a) microbial biomarkers of nonalcoholic fatty liver disease are one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Rikenellaceae, Fusobacteriaceae, Ruminococcaceae, Lachnospiraceae, Actinomycetaceae, Desulfovibrioceae and Desulfovibrionaceae.

    3. The kit according to claim 1, wherein the (a) microbial biomarkers of nonalcoholic fatty liver disease are one or more selected from the group consisting of Citrobacter, Klebsiella, Veillonella, Megamonas, Fusobacterium, Ruminococcus, Faecalibacterium, Oscillospira, Coprococcus, Lachnospira, Actinomyces and Desulfovibrio, wherein the (b) one or more detection markers selected from the group consisting of total bile acid and components of bile acid are one or more selected from the group consisting of total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid, or wherein the (c) intestinal short chain fatty acid is one or more selected from the group consisting of acetate, propionate and butyrate.

    4-5. (canceled)

    6. The kit according to claim 1, wherein the kit comprises one or more selected from combinations of (a) to (p) below: (a) one or more detection means capable of detecting one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Lachnospiraceae and Ruminococcaceae, respectively; (b) one or more detection means capable of detecting one or more selected from the group consisting of total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid, respectively; (c) one or more detection means capable of detecting one or more selected from the group consisting of short chain fatty acid, acetate, propionate and butyrate, respectively; (d) one or more detection means capable of detecting one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Lachnospiraceae, Ruminococcaceae, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid and propionate, respectively; (e) one or more detection means capable of detecting one or more selected from the group consisting of Ruminococcus, Faecalibacterium, Oscillospira, Coprococcus, Lachnospira, total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid, respectively; (f) one or more detection means capable of detecting one or more selected from the group consisting of Ruminococcus, Faecalibacterium, Oscillospira, Coprococcus, Lachnospira and fecal propionate, respectively; (g) one or more detection means capable of detecting one or more selected from the group consisting of Veillonella, Megamonas, total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid, respectively; (h) one or more detection means capable of detecting one or more selected from the group consisting of Veillonella, Megamonas and fecal propionate, respectively, (i) one or more detection means capable of detecting one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Ruminococcaceae, Citrobacter, Klebsiella, Veillonella, Megamonas, Ruminococcus, Faecalibacterium and Oscillospira, respectively; (j) one or more detection means capable of detecting one or more selected from the group consisting of cholic acid, chenodeoxycholic acid, ursodeoxycholic acid and metabolites thereof, respectively; (k) one or more detection means capable of detecting one or more selected from the group consisting of intestinal short chain fatty acid and propionate, respectively, (l) a detection means capable of detecting Enterobacteriaceae, Veillonellaceae, and Ruminococcaceae, (m) a detection means capable of detecting Megamonas and Ruminococcus, (n) a detection means capable of detecting cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, and propionate, (o) combination of (l) and (n), or (p) combination of (m) and (n).

    7-11. (canceled)

    12. The kit according to claim 1, wherein the nonalcoholic fatty liver disease is nonalcoholic fatty liver, nonalcoholic steatohepatitis, liver fibrosis or cirrhosis.

    13. The kit according to claim 1, wherein the nonalcoholic fatty liver disease is non-obese nonalcoholic fatty liver disease, or wherein the kit is for a non-obese patient.

    14. The kit according to claim 1, wherein the predicting the degree of risk is predicting the severity of fibrosis comprising F=0 to F=4.

    15. (canceled)

    16. A method for providing information for predicting or diagnosing a degree of risk of nonalcoholic fatty liver disease comprising detecting one or more detection markers selected from the group consisting of (a) one or more detection markers selected from the group consisting of microbial biomarkers of nonalcoholic fatty liver disease; (b) one or more detection markers selected from the group consisting of total bile acid and components of bile acid; and (c) intestinal short chain fatty acid.

    17. The method according to claim 16, wherein the (a) microbial biomarkers of nonalcoholic fatty liver disease are one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Rikenellaceae, Fusobacteriaceae, Ruminococcaceae, Lachnospiraceae, Actinomycetaceae, Desulfovibrioceae and Desulfovibrionaceae.

    18. The method according to claim 16, wherein the (a) microbial biomarkers of nonalcoholic fatty liver disease are one or more selected from the group consisting of Citrobacter, Klebsiella, Veillonella, Megamonas, Fusobacterium, Ruminococcus, Faecalibacterium, Oscillospira, Coprococcus, Lachnospira, Actinomyces, and Desulfovibrio, wherein the (b) one or more detection markers selected from the group consisting of total bile acid and components of bile acid are one or more selected from the group consisting of total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid, or wherein the (c) intestinal short chain fatty acid is one or more selected from the group consisting of acetate, propionate and butyrate.

    19-20. (canceled)

    21. The method according to claim 16, wherein the method comprises one or more steps selected from combinations of (a) to (p) below: (a) comparing abundances of one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Lachnospiraceae and Ruminococcaceae measured in a subject, with a reference value of a healthy individual; (b) comparing the contents in feces of one or more selected from the group consisting of total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid measured in a subject, with a reference value of a healthy individual; (c) comparing the contents in feces of one or more selected from the group consisting of short chain fatty acid, acetate, propionate and butyrate measured in a subject, with a reference value of a healthy individual; (d) comparing abundances or the contents in feces of one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Lachnospiraceae, Ruminococcaceae, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid and propionate measured in a subject, with a reference value of a healthy individual; (e) comparing abundances or the contents in feces of one or more selected from the group consisting of Ruminococcus, Faecalibacterium, Oscillospira, Coprococcus, Lachnospira, total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid measured in a subject, with a reference value of a healthy individual; (f) comparing abundances or the contents in feces of one or more selected from the group consisting of Ruminococcus, Faecalibacterium, Oscillospira, Coprococcus, Lachnospira and feces propionate measured in a subject, with a reference value of a healthy individual; (g) comparing abundances or the contents in feces of one or more selected from the group consisting of Veillonella, Megamonas, total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid measured in a subject, with a reference value of a healthy individual; (h) comparing abundances or the contents in feces of one or more selected from the group consisting of Veillonella, Megamonas and feces propionate measured in a subject, with a reference value of a healthy individual; (i) comparing abundances of one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Ruminococcaceae, Citrobacter, Klebsiella, Veillonella, Megamonas, Ruminococcus, Faecalibacterium and Oscillospira measured in a subject, with a reference value of a healthy individual; (j) comparing the contents in feces of one or more selected from the group consisting of cholic acid, chenodeoxycholic acid, ursodeoxycholic acid and metabolites thereof measured in a subject, with a reference value of a healthy individual; (k) comparing the contents in feces of one or more selected from the group consisting of intestinal short chain fatty acid and propionate measured in a subject, with a reference value of a healthy individual, (l) comparing abundances of Enterobacteriaceae, Veillonellaceae, and Ruminococcaceae measured in a subject, with a reference value of a healthy individual; (m) comparing abundances of Megamonas and Ruminococcus measured in a subject, with a reference value of a healthy individual; (n) comparing the contents in feces of cholic acid, chenodeoxycholic acid, ursodeoxycholic acid and propionate measured in a subject, with a reference value of a healthy individual; (o) comprising (l) and (n); and (p) comprising (m) and (n).

    22-25. (canceled)

    26. The method according to claim 16, wherein the nonalcoholic fatty liver disease is nonalcoholic fatty liver, nonalcoholic steatohepatitis, liver fibrosis or cirrhosis.

    27. The method according to claim 16, wherein the nonalcoholic fatty liver disease is non-obese nonalcoholic fatty liver disease, or wherein the method is for a non-obese patient.

    28. The method according to claim 16, wherein the method further comprises treating the subject determined to have risk of nonalcoholic fatty liver disease.

    29. A method for screening a therapeutic agent for nonalcoholic fatty liver disease, comprising (1) administering a test substance to an experimental animal having nonalcoholic fatty liver disease; (2) measuring one or more of detection markers selected from the group consisting of (a) one or more detection markers selected from the group consisting of microbial biomarkers of nonalcoholic fatty liver disease; (b) one or more detection markers selected from the group consisting of total bile acid and components of bile acid; and (c) one or more detection markers selected from the group consisting of intestinal short chain fatty acids, in an experimental animal untreated with the test substance and the experimental animal administered with the test substance; and (3) comparing the measured results in a control group untreated with the test substance and the experimental animal administered with the test substance.

    30. The method according to claim 29, wherein the (a) microbial biomarkers of nonalcoholic fatty liver disease are one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Rikenellaceae, Fusobacteriaceae, Ruminococcaceae, Lachnospiraceae, Actinomycetaceae, Desulfovibrioceae and Desulfovibrionaceae.

    31. The method according to claim 29, wherein the (a) microbial biomarkers of nonalcoholic fatty liver disease are one or more selected from the group consisting of Citrobacter, Klebsiella, Veillonella, Megamonas, Fusobacterium, Ruminococcus, Faecalibacterium, Oscillospira, Coprococcus, Lachnospira, Actinomyces, and Desulfovibrio, wherein the (b) one or more detection markers selected from the group consisting of total bile acid and components of bile acid are total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid, and deoxycholic acid, or wherein the (c) intestinal short chain fatty acid is one or more selected from the group consisting of acetate, propionate and butyrate.

    32-33. (canceled)

    34. The method for screening a therapeutic agent of nonalcoholic fatty liver according to claim 25, wherein the method comprises one or more steps selected from combinations of (a) to (p) below: (a) comparing abundances of one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Lachnospiraceae and Ruminococcaceae, measured before and after administration of a candidate substance of therapeutic agent; (b) comparing the contents in feces of one or more selected from the group consisting of total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid, measured before and after administration of a candidate substance of therapeutic agent; (c) comparing the contents in feces of one or more selected from the group consisting of short chain fatty acid, acetate, propionate and butyrate, measured before and after administration of a candidate substance of therapeutic agent; (d) comparing abundances or the contents in feces of one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Lachnospiraceae, Ruminococcaceae, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid and propionate, measured before and after administration of a candidate substance of therapeutic agent; (e) comparing abundances or the contents in feces of one or more selected from the group consisting of Ruminococcus, Faecalibacterium, Oscillospira, Coprococcus, Lachnospira, total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid, measured before and after administration of a candidate substance of therapeutic agent; (f) comparing abundances or the contents in feces of one or more selected from the group consisting of Ruminococcus, Faecalibacterium, Oscillospira, Coprococcus, Lachnospira and feces propionate, measured before and after administration of a candidate substance of therapeutic agent; (g) comparing abundances or the contents in feces of one or more selected from the group consisting of Veillonella, Megamonas, total bile acid, cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, lithocholic acid and deoxycholic acid, measured before and after administration of a candidate substance of therapeutic agent; (h) comparing abundances or the contents in feces of one or more selected from the group consisting of Veillonella, Megamonas and feces propionate, measured before and after administration of a candidate substance of therapeutic agent; (i) comparing abundances of one or more selected from the group consisting of Enterobacteriaceae, Veillonellaceae, Ruminococcaceae, Citrobacter, Klebsiella, Veillonella, Megamonas, Ruminococcus, Faecalibacterium and Oscillospira, measured before and after administration of a candidate substance of therapeutic agent; (j) comparing the contents in feces of one or more selected from the group consisting of cholic acid, chenodeoxycholic acid, ursodeoxycholic acid and metabolites thereof, measured before and after administration of a candidate substance of therapeutic agent; (k) comparing the contents in feces of one or more selected from the group consisting of intestinal short chain fatty acids and propionate, measured before and after administration of a candidate substance of therapeutic agent; (l) comparing abundances of Enterobacteriaceae, Veillonellaceae, and Ruminococcaceae, measured before and after administration of a candidate substance of therapeutic agent; (m) comparing abundances of Megamonas and Ruminococcus, measured before and after administration of a candidate substance of therapeutic agent; (n) comparing the contents in feces of cholic acid, chenodeoxycholic acid, ursodeoxycholic acid and propionate, measured before and after administration of a candidate substance of therapeutic agent; (o) comprising the (l) and (n); and (p) comprising the (m) and (n).

    35. (canceled)

    36. The method for screening a therapeutic agent for nonalcoholic fatty liver disease according to claim 29, wherein the nonalcoholic fatty liver disease is non-obese nonalcoholic fatty liver disease.

    37-42. (canceled)

    43. The method according to claim 16, wherein the method further comprises administering a therapeutic agent of nonalcoholic fatty to the subject determined to have risk of nonalcoholic fatty liver disease.

    Description

    BRIEF DESCRIPTION OF THE DRAWINGS

    [0085] FIG. 1a to 1l are results of comparing the diversities of the gut microbiome. Values dividing the alpha and beta diversity by histological spectrum of NAFLD or fibrosis severity of all subjects are shown in FIG. 1a to 1d, of non-obese subjects are shown in FIG. 1e to 1h, and of obese subjects are shown in FIG. 1i to FIG. 1l. Rarefaction curves were generated using the Shannon index with 12,000 sequences per sample. Statistical analysis was performed using nonparametric Kruskal-Wallis test. NMDS plots were generated using relative OTU abundance data according to Bray-Curtis distance, and statistical significance was determined using Adonis analysis. **P<0.01

    [0086] FIG. 2a to 2d are univariate analysis results for differences in specific microbial taxa according to the severity of fibrosis. For clarity, 13 family- and 14 genus-level taxa are shown along with their upper relative abundance. The box plot shows the interquartile range (IQR) between the first and third quartiles with Tukey whiskers. The color in the box indicates the severity of fibrosis. For statistical significance, a nonparametric Kruskal-Wallis test was used. *P<0.05, **P<0.01, ***P<0.001

    [0087] FIG. 2e to 2h are multivariate analysis results for differences in specific microbial taxa according to the severity of fibrosis. Arcsine root-modified abundance of four bacteria was regressed for age, gender and BMI according to the severity of fibrosis, and the standard residual was indicated as a box plot. *P<0.05, **P<0.01, ***P<0.001

    [0088] FIG. 2i to 2k are results of co-expression analysis of specific gut microbiota elements in total, non-obese and obese subjects. Solid lines (orange) and dotted lines (grey) indicate positive and negative correlations, respectively. The size of the node indicates the relative amount of bacteria, and the color indicates the degree of correlation according to the severity of fibrosis.

    [0089] FIG. 3a to 3j are evaluation results of fecal metabolites mainly related to the gut microbiota. FIG. 3a is the bile acid profile result in various clinical environments, and stacked plots were generated using the average abundance of five bile acids. In FIG. 3b to 3g, the box plots show stratified fecal bile acid levels according to fibrosis severity and obesity status. The concentrations of the five fecal bile acids were stratified by fibrosis severity and obesity status. In FIG. 3h to FIG. 3j, the box plots show the most abundant fecal SCFA (acetate, propionate and butyrate) stratified by fibrosis severity and obesity status. The interquartile ranges (IQRs) between the first and third quartiles are described as Tukey whiskers. For statistical significance, a nonparametric Kruskal-Wallis test was used. *P<0.05, **P<0.01, ***P<0.001

    [0090] FIG. 4 is the network profile result between microbial taxa and fecal metabolite components in the non-obese (a) and obese (b) subjects. Co-expression coefficients between family-level microbiota elements and fecal metabolites were calculated using SparCC and described using Cytoscape. Solid lines (orange) and dotted lines (grey) indicate positive and negative correlations, respectively. The shape of the node indicates the components used in the present study (oval: microbiota, diamond: fecal bile acid, and round rectangle: SCFA), and the color indicates the degree of correlation according to the severity of fibrosis.

    [0091] FIG. 5a and FIG. 5b are receiver operating characteristic curves (ROC) for prediction of significant fibrosis in total, non-obese and obese subjects. FIG. 5a is the ROC curve using the combination of three selected bacteria (Veillonellaceae, Ruminococcacea, and Enterobacteriaceae) and four fecal metabolites (CD, CDCA, UDCA, and propionate) drawn for prediction of significant fibrosis in all the non-obese subjects and obese subjects, and the areas under the curve (AUC) was calculated. FIG. 5b is the ROC curve using the combination of two selected bacteria (Megamonas and Ruminococcus) and four fecal metabolites (CD, CDCA, UDCA, and propionate) drawn for prediction of significant fibrosis in all the non-obese subjects and obese subjects, and the areas under the curve (AUC) was calculated.

    [0092] FIG. 6 is a schematic diagram that comprehensively summarizes the contents corresponding to the differences in the gut microbiome, changes in microorganisms and metabolites, and the prediction result of fibrosis through the combination of representative microorganisms and metabolites, shown in the non-obese subjects and obese subjects.

    [0093] FIG. 7 is a diagram showing the histological distribution of study subjects stratified by fibrosis severity.

    [0094] FIG. 8 is a diagram showing the correlation between the microbial taxa and metabolic indexes in the non-obese and obese patients.

    [0095] FIG. 9a to 9e are diagrams showing the correlation between the microbial taxa and metabolic indexes in the non-obese and obese patients.

    [0096] FIG. 10a to 10c are diagrams showing the relationship between the relative abundance of specific gut microbial taxa and the severity of fibrosis at the genus level stratified by the degree of obesity.

    [0097] FIG. 11 is a diagram showing the relationship between the relative abundance of Actiomyces stratified by the degree of obesity and the TM6SF2 (rs58542926) variant.

    [0098] FIG. 12a to 12d are diagrams showing the relationship between the relative abundance of specific gut microbial components and the presence or absence of diabetes mellitus.

    [0099] FIG. 13a to 13e are diagrams showing the relative abundance of fecal bile acid stratified by the severity of fibrosis and the degree of obesity.

    MODE FOR INVENTION

    [0100] Hereinafter, specific Examples are provided to help the understanding of the present invention, but the following Examples are only illustrative of the present invention, and it is apparent to those skilled in the art that various changes and modifications are possible within the scope and spirit of the present invention, and it is also obvious that these changes and modifications fall within the scope of the appended claims. In the following Examples and comparative Examples, “%” and “part” indicating the content are by weight unless otherwise specified.

    [0101] The values presented in the following experimental Example are expressed as means±standard deviation (S.D.), and the statistical significance of the difference between each treatment group was determined by one-way ANOVA using Graph Pad Prism 4.0 (San Diego. Calif.).

    Experimental Example 1

    [0102] 1. Material and Method

    [0103] 1) Experimental Subject

    [0104] 171 subjects demonstrated by biopsy to have NAFLD and 31 subjects without NAFLD were included. When NAFLD was confirmed histologically and BMI was BMI<25 kg/m.sup.2, it was classified as the non-obese NAFLD group.

    [0105] 2) Subject Inclusion and Exclusion Criteria

    [0106] Subjects were enrolled long-term from January 2013 to February 2017, and the inclusion criteria were as follows:

    [0107] 1. An adult at least 18 years of age,

    [0108] 2. Ultrasonic findings confirming fatty infiltration of liver, and

    [0109] 3. An increase of alanine aminotransferase (ALT) level of unknown etiology within the past 6 months.

    [0110] On the other hand, subjects who met any of the following criteria were excluded:

    [0111] 1. Hepatitis B or C infection,

    [0112] 2. Autoimmune hepatitis, primary biliary cholangitis, or primary sclerosing cholangitis,

    [0113] 3. Gastrointestinal cancers or hepatocellular carcinoma,

    [0114] 4. Drug-induced steatosis or liver damage,

    [0115] 5. Wilson disease or hemochromatosis,

    [0116] 6. Excessive alcohol consumption (male: >210 g/week, female: >140 g/week),

    [0117] 7. Antibiotic use within the previous month,

    [0118] 8. Diagnosis of malignancy in the past year,

    [0119] 9. Human immunodeficiency virus infection, and

    [0120] 10. Chronic disorders related to lipodystrophy or immunosuppression.

    [0121] Non-obese and obese control groups included subjects without any suspicion of NFALD (a) during evaluation of living donor liver transplantation or (b) during liver biopsy for characterization of solid liver mass suspected for hepatic adenoma or focal nodular hyperplasia based on imaging studies (Koo B K, Joo S K, Kim D, Bae J M, Park J H, Kim J H, et al. Additive effects of PNPLA3 and TM6SF2 on the histological severity of non-alcoholic fatty liver disease. J Gastroenterol Hepatol 2018; 33:1277-1285.).

    [0122] 3) Liver Histology

    [0123] Liver histology was evaluated by a single liver pathologist using the NASH CRN histological scoring system. NAFLD was defined as the presence of ≥5% macrovesicular steatosis based on histological examination. NASH was defined based on the overall pattern of liver damage consisting of steatosis, lobular inflammation or ballooning of hepatocytes according to the criteria of Brunt et al. (Brunt E M, Janney C G, Di Bisceglie A M, Neuschwander-Tetri B A, Bacon B R. Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions. Am J Gastroenterol 1999; 94:2467-2474; Brunt E M, Kleiner D E, Wilson L A, Belt P, Neuschwander-Tetri B A. Nonalcoholic fatty liver disease (NAFLD) activity score and the histopathologic diagnosis in NAFLD: distinct clinicopathologic meanings. Hepatology 2011; 53:810-820). In addition, steatosis, hepatic lobular inflammation and swelling were scored according to the NAFLD activity scoring system, and the severity of fibrosis was evaluated according to the criteria of Kleiner et al. (Kleiner D E, Brunt E M, Van Natta M, Behling C, Contos M J, Cummings O W, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 2005; 41:1313-1321).

    [0124] 4) Microbiome Analysis Using 16S rRNA Sequencing

    [0125] DNA of the fecal sample was extracted using QIAamp DNA Stool Mini Kit (Qiagen, Hilden, Germany). V4 region sequencing targeting of 16S rRNA was performed using MiSeq platform (Illumina, San Diego, Calif., USA), and additional treatment of raw sequencing data was performed using QIIME pipeline (v 1.8.0) (Caporaso J G, Kuczynski J, Stombaugh J, Bittinger K, Bushman F D, Costello E K, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods 2010; 7:335-336).

    [0126] 5) Measurement of Fecal Metabolites Using GC-FID and Q-TOP System

    [0127] Fecal SCFA was measured using Agilent Technologies 7890A GC system (Agilent Technologies, Santa Clara, Calif., USA) according to the method of David (David L A, Maurice C F, Carmody R N, Gootenberg D B, Button J E, Wolfe B E, et al. Diet rapidly and reproducibly alters the human gut microbiome. Nature 2014; 505:559-563), and the bile acid profile was evaluated using Q-TOF mass spectrometer (Waters Micromass Technologies, Manchester, UK).

    [0128] 6) Bioinformatics Analysis and Statistical Test

    [0129] Statistical comparison was performed with Kruskal-Wallis test using GraphPad Prism software Ver. 7.0d (GraphPad Software, San Diego, Calif., USA). For rarefaction curves, the OUT table was selected by 12,000 sequences per sample, and Shannon index was measured by QIIME. Nonparametric multi-dimensional scaling (NMDS) plots were represented using Vergan package of R (Oksanen J, Kindt R, Legendre P, O'Hara B, Stevens M H H, Oksanen M J, et al. The vegan package. Community ecology package 2007; 10.), and the distance was measured using Bray-Curtis method. The statistical significance between groups was estimated using Adonis function. Multivariate association analysis using microbiome data was performed using multivariate association using a linear model (MaAsLin) for identification of specific taxa related to the host phenotype without being affected by other metadata (Morgan X C, Tickle T L, Sokol H, Gevers D, Devaney K L, Ward D V, et al. Dysfunction of the gut microbiome in inflammatory bowel disease and treatment. Genome Biol 2012; 13:R79.). In addition, age, gender and BMI or diabetes were designated as fixed variables, and when the p-value adjusted by Benjamini and Hochberg's false discovery rate (FDR) was lower than 0.20, the association rate was considered as significant.

    [0130] 7) Significant Prediction of Fibrosis by ROC Curves

    [0131] In order to demonstrate the prediction ability of fibrosis of the microbiome-based biomarkers, the area under the receiver operating characteristic curve (AUROC) method was used. The three family-level bacteria, basic characteristics of subjects (age, gender and BMI) and relative abundance of FIB-4 confirmed in the present experiment were used as inputs for AUROC, and the combination of their factors was calculated using binary logistic regression in SPSS Ver. 25.0 (SPSS Inc., Armonk, N.Y., USA). AUROC comparison was performed by DeLong test using MedCalc software Ver. 18.2.1 (MedCalc Software BVBA, Ostend, Belgium).

    [0132] 2. Experimental Result

    [0133] 1) Basic Characteristics

    [0134] 171 subjects demonstrated as NAFLD (NAFL, n=88; NASH, n=83) by biopsy and 31 non-NAFLD subjects were included, and all subjects were divided into two groups (non-obese, BMI<25; obese, BMI≥25), and each subject was divided into three subgroups according to the histological spectrum of NAFLD or fibrosis. In Table 1 and Table 2, the result of detailed characteristics of each group including clinical, metabolic, biochemical and histological profiles was shown.

    TABLE-US-00001 TABLE 1 Baseline characteristics of study subjects stratified by obesity status and histological spectrum of NAFLD. Non-obese (n = 64) Obese (n = 138) No No NAFLD NAFL NASH P-value NAFLD NAFL NASH P-value N 7/14 13/11 7/12 4/6 37/27 24/40 (male/female) Age 58.7 ± 10.7 58.3 ± 10.2 60.2 ± 8.84 0.8601 .sup.ns   58 ± 12.6 52.7 ± 14.8 53.6 ± 16.7  0.6463 .sup.ns (years) BMI 22.8 ± 1.67 23.6 ± 1.34 23.6 ± 0.83 0.0871 .sup.ns   27 ± 2.09 28.8 ± 3.25 28.8 ± 3.02  0.1374 .sup.ns (kg/m.sup.2) WC (cm)  80 ± 6.53.sup.a  82.7 ± 3.45.sup.ab 85.4 ± 4.sup.b   0.0141 *  92.7 ± 5.93 94.3 ± 8.04 96.6 ± 7.81  0.2328 .sup.ns AST  31.6 ± 24.4.sup.a 28.4 ± 9.05.sup.b   54.7 ± 45.7.sup.bc 0.002 **  25.6 ± 7.5.sup.a  42.3 ± 26.5.sup.b   62 ± 32.3.sup.bc <0.0001 *** (IU/L) ALT  32.5 ± 32.6.sup.a 32.8 ± 17.3.sup.b   59 ± 52.9.sup.c <0.0001 ***   27.8 ± 25.8.sup.a  56.9 ± 48.5.sup.b 79.1 ± 57.6.sup.c <0.0001 *** (IU/L) GGT 44.6 ± 54.sup.a  31.8 ± 34.1.sup.a  66.7 ± 55.2.sup.b 0.0016 **  44.7 ± 51.8.sup.a  49.2 ± 57.4.sup.a 78.5 ± 79.2.sup.b <0.0001 *** (IU/L) HDL 54 ± 14 46.3 ± 10.8 43.8 ± 11.7 0.0527 .sup.ns  58.9 ± 14.3.sup.a  45.5 ± 11.6.sup.b 45.5 ± 11.2.sup.bc 0.0196 * cholesterol (mg/dL) LDL 93.5 ± 25.7 111 ± 39.8 97.3 ± 32.1 0.5322 .sup.ns  123 ± 35.8 103 ± 32.3 107 ± 32.7 0.1782 .sup.ns cholesterol (mg/dL) Albumin  4.1 ± 0.285  4.24 ± 0.257  4.03 ± 0.413 0.1128 .sup.ns   4.12 ± 0.312   4.2 ± 0.254 4.15 ± 0.279 0.516 .sup.ns (g/dL) Platelet   223 ± 72.2.sup.ab  259 ± 54.4.sup.a 179 ± 79.sup.bc 0.0023 ** 232 ± 53  234 ± 59.9 215 ± 69  0.3027 .sup.ns (×10.sup.3/μL) Ferritin 103 ± 57.2 109 ± 80.2 173 ± 114 0.1285 .sup.ns   63.8 ± 26.9.sup.a  137 ± 88.4.sup.b  159 ± 95.1.sup.bc 0.0026 ** (ng/mL) HA  66.6 ± 70.2.sup.ab 37.1 ± 30.2.sup.a  93.9 ± 64.9.sup.bc 0.0048 **   76.8 ± 99.2.sup.ab  62.1 ± 95.4.sup.a 95.7 ± 112.sup.bc 0.0344 * (ng/mL) Insulin 9.45 ± 3.77.sup.a  11.2 ± 5.96.sup.ab  13.6 ± 5.85.sup.b 0.037 *   11.2 ± 5.55.sup.a  18.1 ± 17.1.sup.ab 18.2 ± 11.1.sup.b 0.0227 * (μIU/mL) HbA1c (%)  5.71 ± 0.481.sup.a  6.06 ± 0.676.sup.a  7.12 ± 1.96.sup.b <0.0001 ***    5.72 ± 0.326.sup.a    6.1 ± 0.814.sup.ab  6.6 ± 1.27.sup.b 0.0099 ** C-peptide 1.91 ± 0.61.sup.a  2.43 ± 0.832.sup.ab  2.88 ± 1.09.sup.b  0.0005 ***    2.42 ± 0.916.sup.a 4.42 ± 3.4.sup.b 4.19 ± 2.39.sup.bc 0.0087 ** (ng/mL) HOMA-IR 2.56 ± 1.17.sup.a  3.12 ± 1.72.sup.ab  4.39 ± 2.19.sup.bc 0.0085 **   2.92 ± 1.77.sup.a  4.96 ± 4.36.sup.ab 5.77 ± 4.42.sup.bc 0.0131 * Adipo-IR 4.68 ± 2.85.sup.a   6.6 ± 3.59.sup.ab  9.63 ± 5.39.sup.bc 0.0027 **   6.42 ± 3.28.sup.a  10.1 ± 10.1.sup.ab 12.8 ± 9.51.sup.bc 0.0096 ** FFA (μEq/L)  493 ± 166.sup.ab 620 ± 240.sup.a  720 ± 268.sup.bc 0.0121 *   605 ± 289a  603 ± 259.sup.ab 712 ± 238.sup.bc 0.0089 ** hsCRP 0.249 ± 0.428.sup.a 0.0896 ± 0.066.sup.a  0.354 ± 0.549.sup.b 0119 *   0.152 ± 0.154.sup.ab 0.206 ± 0.403.sup.a 0.278 ± 0.336.sup.bc 0.0294 * (mg/dL) Cholesterol 167 ± 28.2 185 ± 41.8 167 ± 42.9 0.3932 .sup.ns   200 ± 43.2 183 ± 34.8 181 ± 40.7  0.4122 .sup.ns (mg/dL) TG 102 ± 47.sup.a   140 ± 44.7.sup.b  141 ± 70.9.sup.ab 0.0105 *   87 ± 34.sup.a  161 ± 82.2.sup.b 151 ± 61.9.sup.bc 0.0024 ** (mg/dL) FPG 110 ± 25.6 113 ± 28.8 132 ± 42.5 0.0986 .sup.ns   102 ± 14.1  113 ± 27.4 128 ± 55.8  0.2074 .sup.ns (mg/dL) HTN, 8 (38.1) 8 (33.3) 9 (47.4) 0.641 .sup.ns 4 .sup.ns 24 (37.5) 32 (50.0) 0.352 n (%) Diabetes, 1 (4.76) 8 (33.3) 13 (68.4) 0.0001 *** 1 * 19 (29.7) 30 (46.9) 0.026 n (%) Abbreviations: BMI, body mass index; WC, waist circumference; AST, aspartatetransaminase; ALT, alanine transaminase; GGT, gamma-glutamyl transferase; NAS, nonalcoholic fatty liver disease activity score; HDL, high-density lipoprotein; LDL, low-density lipoprotein; HA, hyaluronic acid; HbAlc, glycosylated hemoglobin; HOMA-IR, homeostasis model assessment of insulin resistance; Adipo-IR, adipose tissue insulin resistance; PM, free fatty acid; hsCRP, high-sensity C-reactive protein; TG, triglycerides; FBG, fasting blood glucose; HTN, hypertension. Data are expressed as the mean ± SD or n (%). Mean ± SD or n (%) with defferent superscript letters indicates significant differences by the nonparametric Kruskal-Wallis test or the chi-square test. *P < 0.05, **P < 0.01, ***P < 0.001

    TABLE-US-00002 TABLE 2 Baseline characteristics of study subjects stratified by obesity status and fibrosis severity. Non-obese (n = 64) Fibrosis stage 0 1 ≥2 P-value N (male/female) 27 (11/16) 20 (9/11) 17 (7/10) Age (years) 57.67 ± 9.01 57.85 ± 11.80 62.47 ± 8.32 0.2371 .sup.ns BMI (kg/m.sup.2)  22.81 ± 1.42.sup.a 23.76 ± 1.46.sup.b  23.71 ± 0.92.sup.ab 0.0084 ** WC (cm) 79.95 ± 5.2.sup.a  83.22 ± 3.53.sup.ab 86.33 ± 4.22.sup.b 0.0010 ** AST (IU/L)   29.11 ± 21.64.sup.a   32.20 ± 10.63.sup.ab  56.06 ± 48.13.sup.b 0.0017 ** ALT (IU/L)   31.89 ± 27.83 39.10 ± 30.99  55.71 ± 52.04 0.0886 .sup.ns GGT (IU/L)   33.6 ± 41.4.sup.a   44 ± 46.2.sup.ab   69.7 ± 58.3.sup.b 0.0046 ** HDL-cholesterol  51.4 ± 13.4 47.9 ± 11.1  43.1 ± 12.3 0.1701 .sup.ns (mg/dL) LDL-cholesterol  104 ± 29.9  108 ± 38.6  90.8 ± 32.3 0.3250 .sup.ns (mg/dL) Albumin (g/dL)   4.14 ± 0.25  4.22 ± 0.29  3.99 ± 0.43 0.1781 .sup.ns Platelet (×10.sup.3/μL)  230.19 ± 48.76.sup.a 247.75 ± 74.84.sup.a 183.88 ± 95.35.sup.b 0.0255 * Ferritin (ng/mL) 117.75 ± 73.97 100.94 ± 74.93   282.19 ± 386.91 0.053 .sup.ns HA (ng/mL)  33.08 ± 19.62.sup.a  64.2 ± 67.47.sup.a 109.59 ± 65.56.sup.b 0.0002 *** Insulin (μIU/mL) 10.76 ± 5.57 10.05 ± 4.15 13.71 ± 6.19  0.1103 .sup.ns HbAlc (%)   5.86 ± 0.69.sup.a  5.98 ± 0.44.sup.ab  7.23 ± 2.05.sup.c 0.0007 *** C-peptide (ng/mL)   2.23 ± 0.87.sup.a  2.23 ± 0.64.sup.ab  2.85 ± 1.17.sup.bc 0.0256 * HOMA-IR   2.94 ± 1.61.sup.a  2.81 ± 1.33.sup.ab   4.50 ± 2.28.sup.b 0.0207 * Adipo-IR  5.72 ± 3.01  6.33 ± 3.72  9.49 ± 5.95 0.0645 .sup.ns FFA (μEq/L)  553.96 ± 186.66  615.65 ± 238.13  684.88 ± 308.55 0.2678 .sup.ns hsCRP (mg/dL)   0.17 ± 0.33.sup.a  0.23 ± 0.41.sup.ab  0.29 ± 0.48.sup.bc 0.0186 * Cholesterol (mg/dL) 177.7 ± 28   180.75 ± 43.68 158.53 ± 44.94 0.1860 .sup.ns TG (mg/dL) 120.70 ± 45.23 128.42 ± 51.84 137.12 ± 77.04 0.9889 .sup.ns FPG (mg/dL)  111.15 ± 31.51.sup.a  111.85 ± 19.58.sup.ab  134.47 ± 43.79.sup.b 0.0402 * HTN, n (%) 7 (25.9) 9 (45.0) 9 (52.9) 0.163 .sup.ns Diabetes, n (%) 4 (14.8) 5 (25.0) 13 (76.5) 0.0001 *** Obese (n = 138) Fibrosis stage 0 1 ≥2 P-value N (male/female) 25 (17/8) 73 (38/35) 40 (10/30) Age (years)  57.08 ± 12.41  48.36 ± 15.70.sup.a   60.63 ± 13.56.sup.b 0.0001 *** BMI (kg/m.sup.2)  27.48 ± 2.58.sup.a 29.30 ± 3.20.sup.b  28.27 ± 2.97.sup.ab 0.0119 * WC (cm) 91.82 ± 6.56 96.16 ± 7.35  95.73 ± 8.98 0.0074 .sup.ns AST (IU/L)  28.40 ± 13.03.sup.a  48.99 ± 26.74.sup.b   66.13 ± 36.43.sup.c <0.0001 *** ALT (IU/L)  38.88 ± 43.67.sup.a   70.85 ± 53.02.sup.bc   70.85 ± 56.63.sup.c 0.0003 *** GGT (IU/L)  39.6 ± 41.1.sup.a   57.5 ± 56.1.sup.ab   85.9 ± 95.2.sup.b 0.0016 ** HDL-cholesterol   48.4 ± 13.1  47.2 ± 12    46.3 ± 11.5 0.9175 .sup.ns (mg/dL) LDL-cholesterol  105 ± 35.7  109 ± 31.7   102 ± 33.8 0.5024 .sup.ns (mg/dL) Albumin (g/dL)   4.12 ± 0.24.sup.ab  4.24 ± 0.26.sup.a  4.08 ± 027.sup.b 0.0027 ** Platelet (×10.sup.3/μL)  238.2 ± 55.38.sup.a 241.44 ± 62.49.sup.a  188.33 ± 58.05.sup.b 0.0001 *** Ferritin (ng/mL)  145.55 ± 89.51  219.37 ± 255.97  169.26 ± 133.32 0.8403 .sup.ns HA (ng/mL)  51.32 ± 66.4.sup.a  61.58 ± 90.48.sup.a 127.28 ± 129.4.sup.b 0.0001 *** Insulin (μIU/mL)   14.22 ± 11.57.sup.a   17.83 ± 15.46.sup.ab  19.50 ± 12.58.sup.b 0.0148 * HbAlc (%)   5.87 ± 0.54.sup.a   6.15 ± 0.85.sup.ab  6.87 ± 1.42.sup.c 0.0007 *** C-peptide (ng/mL)   3.22 ± 1.57.sup.a   4.43 ± 3.43.sup.ab   4.29 ± 2.28.sup.bc 0.0498 * HOMA-IR    3.84 ± 3.35.sup.a   4.82 ± 4.18.sup.ac  6.69 ± 4.70.sup.b 0.0006 *** Adipo-IR  7.48 ± 6.2.sup.a 10.76 ± 9.7.sup.ab    13.86 ± 10.55.sup.bc 0.0043 ** FFA (μEq/L)   556.08 ± 209.52.sup.a   642.76 ± 257.61.sup.ab    737.1 ± 259.42.sup.bc 0.0059 ** hsCRP (mg/dL)   0.14 ± 0.17.sup.a  0.23 ± 0.39.sup.b    0.3 ± 0.39.sup.bc 0.0121 * Cholesterol (mg/dL) 180.96 ± 38.04 188.58 ± 34.01  175.33 ± 44.67 0.2143 .sup.ns TG (mg/dL)  127.36 ± 51.42 156.23 ± 80.68  155.43 ± 67.19  0.2363 .sup.ns FPG (mg/dL)  110.96 ± 34.36.sup.a 107.82 ± 21.43.sup.a  144.53 ± 63.53.sup.b 0.0001 *** HTN, n (%) 9 (36.0) 30 (41.1) 21 (52.5) 0.357 .sup.ns Diabetes, n (%) 3 (12.0) 23 (31.5) 24 (60.0) 0.0002 ***

    [0135] As a result of confirmation, subjects with NASH or significant fibrosis (F2-4) had high levels of aspartate aminotransferase (AST), alanine aminotransferase (ALT) and diabetic markers in all obese and non-obese groups. The subjects with significant fibrosis had higher NAFLD activity scores, and showed more severe liver histology in terms of histological classification of NAFLD (Table 3 and FIG. 7). More detailed standard characteristics of each fibrosis stage, comprising well-known NAFLD-related genetic variations such as PNPLA3, TM6SF2, MBOAT7-TMC4, and SREBF-2 were shown in Table 4.

    TABLE-US-00003 TABLE 3 Histological characteristics of study subjects stratified by obesity status and fibrosis severity. Non-obese (n = 64) Obese (n = 138) Fibrosis stage 0 1 2 0 1 2 Steatosis, n (%) 0 (<5%) 15 (55.6) 5 (25.0) 1 (5.9) 7 (28.0) 2 (2.7) 1 (2.5) 1 (5-33%) 7 (25.9) 6 (30.0) 8 (47.1) 12 (48.0) 9 (12.3) 13 (32.5) 2 (34-66%) 4 (14.8) 7 (35.0) 2 (11.8) 3 (12.0) 29 (39.7) 12 (30.0) 3 (>66%) 1 (3.7) 2 (10.0) 6 (35.3) 3 (12.0) 33 (45.2) 14 (35.0) Lobular inflammation, n (%) 0 15 (55.6) 3 (15.0) 1 (5.9) 13 (52.0) 5 (6.9) 3 (7.5) 1 12 (44.4) 14 (70.0) 11 (64.7) 12 (48.0) 60 (82.2) 30 (75.0) 2-3 0 3 (15.0) 5 (29.4) 0 8 (11.0) 7 (17.5) Ballooning, n (%) 0 22 (81.5) 7 (35.0) 0 22 (88.0) 18 (24.7) 5 (12.5) 1-2 5 (18.5) 13 (65.0) 17 (100.0) 3 (12.0) 55 (75.3) 35 (87.5) Histological classification, n (%) No NAFLD 15 (55.6) 5 (25.0) 1 (5.9) 7 (28.0) 2 (2.7) 0 NAFL 11 (40.7) 13 (65.0) 0 18 (72.0) 40 (54.8) 7 (17.5) NASH 1 (3.7) 2 (10.0) 16 (94.1) 0 31 (42.5) 33 (82.5) NAS 1.30 ± 1.46 2.95 ± 1.61 4.00 ± 1.32 1.68 ± 1.22 4.11 ± 1.23 4.08 ± 1.10 Abbreviations: NAFLD, nonalcoholic fatty liver disease; NAFL, nonalcoholic fatty liver; NASH, nonalcoholic steatohepatitis; NAS, NAFLD activity score. Date are expressed as the mean ± SD or n (%).

    TABLE-US-00004 TABLE 4 Baseline clinical, metabolic, histological, and genetic characteristics of study subjects stratified by obesity status and fibrosis stage. Non-obese (n = 64) Fibrosis stage 0 1 2 3 4 N (male/female) 27 (11/16) 20 (9/11) 9 (5/4) 4 (1/3) 4 (1/3) Age (years) 57.7 ± 9.01 57.8 ± 11.8 58.6 ± 9.9  67.2 ± 2.5  66.5 ± 1.73 BMI 22.8 ± 1.42 23.8 ± 1.46  23.5 ± 0.862 23.9 ± 1.33   24 ± 0.66 WC (cm) 79.9 ± 5.2  83.2 ± 3.53 85.5 ± 2.27 88.1 ± 4.3  86.1 ± 8.37 SBP (mm Hg)  128 ± 16.9  127 ± 14.3 132 ± 17.1  158 ± 37.4 124 ± 21.7 DBP (mm Hg) 76.8 ± 12.8 77.4 ± 8.26 80.2 ± 12.9 86.8 ± 19.1 74.5 ± 10.8 AST (IU/L) 29.1 ± 21.6 32.2 ± 10.6 55.1 ± 55.8 80.5 ± 50.1 33.8 ± 8.34 ALT (IU/L) 31.9 ± 27.8 39.1 ± 31   52.6 ± 53.2 93.5 ± 60.5   25 ± 7.53 GGT (IU/L) 33.6 ± 41.4   44 ± 46.2 69.7 ± 64  373 ± 592   63 ± 43.8 Insulin (μIU/mL) 10.8 ± 5.57  10 ± 4.15   15 ± 6.14 15.3 ± 7.18 9.32 ± 4.29 HbAlc (%)  5.86 ± 0.688  5.98 ± 0.445   6.8 ± 0.689  6.05 ± 0.557 9.38 ± 3.52 HOMA-IR 2.94 ± 1.61 2.81 ± 1.33  4.57 ± 2.05 4.56 ± 2.04 4.27 ± 3.5  Adipo-IR 5.72 ± 3.01 6.33 ± 3.72  11.1 ± 6.82 8.94 ± 5.18 6.44 ± 4.2  Diabetes, n (%) 4 (14.8) 5 (25.0) 7 (77.8) 2 (50.0) 4 (100) Albumin (g/dL)  4.15 ± 0.259 4.22 ± 0.291  4.19 ± 0.285 3.92 ± 0.45  3.62 ± 0.499 Platelet (×10.sup.3/μL)  230 ± 48.8 248 ± 74.8  222 ± 67.2 206 ± 123 76.8 ± 32.1 TG (mg/dL) 121 ± 45.2 128 ± 51.8  180 ± 82.9 82.5 ± 26.6   96 ± 31.1 FPG (mg/dL) 111 ± 31.5 112 ± 19.6  123 ± 26.7 123 ± 37   172 ± 66.6 Histological classification No NAFLD 15 (55.6) 5 (25.0) 0 1 (25.0) 0 NAFL 11 (40.7) 13 (65.0) 0 0 0 NASH 1 (3.7) 2 (10.0) 9 (100) 3 (75.0) 4 (100) Genetic variants PNPLA3 G/G 6 (22.2) 4 (20.0) 1 (11.1) 0 2 (50.0) (rs738409) C/G 13 (38.1) 13 (65.0) 5 (55.6) 1 (25.0) 2 (50.0) C/C 7 (25.9) 3 (15.0) 2 (22.2) 3 (75.0) 0 TM6SF2 C/C 21 (77.8) 18 (90.0) 6 (66.7) 2 (50.0) 3 (75.0) (rs58542926) C/T 5 (18.5) 2 (10.0) 2 (22.2) 2 (50.0) 1 (25.0) T/T 0 0 0 0 0 MBOAT7-TMC4 C/C 17 (63.0) 13 (65.0) 5 (55.6) 1 (25.0) 4 (100) (rs641738) C/T 9 (33.3) 5 (25.0) 3 (33.3) 3 (75.0) 0 T/T 0 2 0 0 0 SREBF-2 C/C 7 (25.9) 6 (30.0) 1 (11.1) 2 (50.0) 2 (50.0) (rs133291) C/T 12 (44.4) 8 (40.0) 6 (66.7) 1 (25.0) 2 (50.0) T/T 3 (11.1) 6 (30.0) 1 (11.1) 1 (25.0) 0 Obese (n = 138) Fibrosis stage 0 1 2 3 4 N (male/female) 25 (17/8) 73 (38/35) 20 (5/15) 7 (2/5) 13 (3/10) Age (years) 57.1 ± 12.4 48.4 ± 15.7 55.6 ± 16.4 65.6 ± 8.89 65.8 ± 6.92 BMI 27.5 ± 2.58 29.3 ± 3.2  28.5 ± 3.26  28 ± 3.43   28 ± 2.39 WC (cm) 91.8 ± 6.55 96.2 ± 7.35 95.8 ± 9.02 95.8 ± 13.2 95.6 ± 7.76 SBP (mm Hg) 130 ± 14.8  136 ± 18.4 133 ± 17  132 ± 11.3 126 ±17.9 DBP (mm Hg)  80 ± 9.78 84.1 ± 11.8 77.4 ± 13.4  77 ± 10.8 74.3 ± 9.55 AST (IU/L) 28.4 ± 13     49 ± 26.7 66.6 ± 46.5 70.4 ± 27.9    63 ± 21.7 ALT (IU/L) 38.9 ± 43.7 70.8 ± 53  87.2 ± 74.3  58 ± 19.3  52.6 ± 2 4.6 GGT (IU/L) 39.6 ± 41   57.5 ± 56.1 79.4 ± 86.1 86.6 ± 68.5  95.6 ± 123 Insulin (μIU/mL) 14.2 ± 11.6  17.8 ± 15.5 21.3 ± 16.1 13.5 ± 4.73   20 ± 8.1 HbAlc (%)  5.87 ± 0.538  6.15 ± 0.851 7.01 ± 1.45 6.79 ± 1.15  6.74 ± 1.59 HOMA-IR 3.84 ± 3.36 4.82 ± 4.18 7.28 ± 5.79 4.4 ± 1.42 7.01 ± 3.71 Adipo-IR 7.48 ± 6.2  10.8 ± 9.7  14.3 ± 12.9 9.29 ± 7.57   15.1 ± 7.42 Diabetes, n (%) 3 (12.0) 23 (31.5) 12 (60.0) 4 (57.1) 8 (61.5) Albumin (g/dL)  4.12 ± 0.243 4.24 ± 0.26  4.11 ± 0.192  4.06 ± 0.207  4.04 ± 0.393 Platelet (×10.sup.3/μL) 238 ± 55.4  241 ± 62.5  206 ± 50.6 195 ± 48.8  157 ± 63.7 TG (mg/dL) 127 ± 51.4  156 ± 80.7  175 ± 71.5 140 ± 48.6  134 ± 64.5 FPG (mg/dL) 111 ± 34.4  108 ± 21.4  148 ± 78.3 138 ± 30.6  143 ± 53.9 Histological classification No NAFLD 7 (28.0) 2 (2.7) 0 0 0 NAFL 18 (72.0) 40 (54.8) 6 (30.0) 1 (14.3) 0 NASH 0 31 (42.5) 14 (70.0) 6 (85.7) 13 (100) Genetic variants PNPLA3 G/G 4 (27.4) 20 (27.4) 11 (55.0) 2 (28.6) 6 (46.2) (rs738409) C/G 11 (44.0) 35 (47.9) 4 (20.0) 4 (57.1) 4 (30.8) C/C 7 (17.8) 13 (17.8) 4 (20.0) 1 (14.3) 1 (7.7) TM6SF2 C/C 18 (72.0) 56 (76.7) 16 (80.0) 4 (57.1) 10 (76.9) (rs58542926) C/T 4 (16.0) 11 (15.1) 3 (15.0) 2 (28.6) 1 (7.7) T/T 0 1 (1.4) 0 1 0 MBOAT7-TMC4 C/C 15 (60.0) 42 (57.5) 11 (55.0) 3 (42.9) 7 (53.8) (rs641738) C/T 4 (16.0) 21 (28.8) 8 (40.0) 4 (57.1) 4 (30.8) T/T 3 (12.0) 5 (6.8) 0 0 0 SREBF-2 C/C 6 (24.0) 23 (31.5) 8 (40.0) 1 (14.3) 3 (23.1) (rs133291) C/T 7 (28.0) 27 (37.0) 5 (25.0) 5 (71.4) 5 (38.5) T/T 3 (12.0) 8 (11.0) 4 (20.0) 1 (14.3) 2 (15.4) Abbreviations: BMI, body mass index; WC, waist circumference; SBP, systolic bloodpressure; DBP, diastolic blood pressure; AST, aspartate transaminase; ALT, alaninetransaminase; GGT, gamma-glutamyl transferase; HbAlc, glycosylated hemoglobin;HOMA-IR, homeostasis model assessment of insulin resistance; Adipo-IR, adiposetissue insulin resistance; TG, triglycerides; FBG, fasting blood glucose; NAFLD,nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; PNPLA3, patatin-like phospholipase domain-containing protein 3; TM6SF2, transmembrane 6 superfamily 2; MBOAT7-TMC4, membrane bound O-acyltransferase domain-containing 7 gene and transmembrane channel-like 4 gene; SREBF-2, sterolregulatory element binding transcription factor 2. Data are expressed ± SD or nas mean(%).

    [0136] 2) Observation of Changes in Microbiome According to Fibrosis Severity

    [0137] Depending on the fibrosis severity, changes of the microbiome were shown differently in the non-obese NAFLD subjects and obese NAFLD subjects.

    [0138] Specifically, the microbial diversity was compared according to the histological spectrum of NAFLD or fibrosis severity (FIG. 1). For comparison of alpha diversity, rarefaction curves based on Shannon metric were plotted, and NMDS plots based on Bray-Curtis distance were plotted for beta diversity. As a result of confirmation, any significant changes between groups stratified by the histological spectrum of NAFLD or fibrosis severity were not found in the merged subjects (FIG. 1a to 1d).

    [0139] The subjects were classified into two groups according to their BMI status. In the non-obese group, a significant decrease in microbial diversity was observed between F1 and F0 (p=0.0074), as well as between F2-4 and F0 (p=0.0084) (FIG. 1e to 1h). Moreover, clear clustering between F0 and F2-4 was observed (p=0.038). In the obese group, there was no significant change in diversity between groups stratified by the histological classification of NAFLD or fibrosis severity (FIG. 1i to 1l).

    [0140] The result indicates that the fibrosis severity is more related to gut microbiome change than necroinflammatory activity, and basic BMI status may also be an important factor contributing to gut microbiome change.

    [0141] 3) Observation of Proliferation of Fibrosis-Related Microbial Taxa

    [0142] Proliferation of the fibrosis-related microbial taxa was remarkably shown in the non-obese NAFLD subjects. Specifically, in the non-obese and obese subjects, the differences of the specific microbial taxa according to the fibrosis severity were compared using univariate and multivariate analyses (FIGS. 2a to 2d and 2e to 2h).

    [0143] In the univariate analysis, not only gradual proliferation of Veillonellaceae mostly found in the oral cavity and small intestine and large intestine, but also Enterobacteriaceae were observed according to the fibrosis severity of the non-obese subjects. In the obese subjects, Rikenellaceae became gradually enriched. On the contrary, the abundance of Ruminococcaceae was significantly reduced as fibrosis became more severe, and this was found only in the non-obese subjects. This result could be confirmed in correlation plots (FIG. 8), and Enterobacteriaceae and Veillonellaceae showed a positive correlation with the fibrosis severity (p=1.09×10.sup.−4, p=2.44×10.sup.−3, respectively), but Ruminococcaceae showed an inverse correlation.

    [0144] At the genus level, Faecalibacterium (Ruminococcaceae), Ruminococcus (Ruminococcaceae), Coprococcus (Lachnospiraceae), and Lachnospira (Lachnospiraceae) were significantly drastically reduced in the significant fibrosis group, but the abundance of Enterobacteriaceae_Other (Enterobacteriaceae) and Citrobacter was gradually increased according to the fibrosis severity. This change was observed only in the non-obese subjects.

    [0145] For multivariate analysis, the age, gender and BMI were adjusted using MaAsLin. Enterobacteriaceae was an abundant family significantly related to the fibrosis severity in the non-obese subjects (p=0.0108, q=0.214) (FIG. 2e to 2h). In phylum Firmicutes, Veillonellaceae showed a steep increase of the relative abundance in the non-obese subjects than the obese subjects (non-obese, p=0.0002, q=0.0195), but the abundance of Ruminococcaceae showed an inverse correlation with the fibrosis severity in the non-obese subjects (p=0.0019, q=0.0908). A representative genus of Ruminococcaceae, Ruminococcus also showed a significant inverse correlation according to the fibrosis severity (p=0.0009, q=0.135) (FIG. 10a to 10c). In addition, Veillonellaceae and Enterobacteriaceae showed a significant positive correlation with the serum free fatty acid (FFA) level in the non-obese subjects (q=0.178, q=0.118, respectively), but it did not in the obese subjects (FIG. 9a to 9e).

    [0146] Adipo-IR and glycosylated hemoglobin (HbA1c) also showed a positive correlation according to the abundance of Veillonellaceae (adipo-IR, q=0.142; HbA1c, q=0.157). On the contrary, the serum FFA level showed an inverse correlation with the abundance of Ruminococcus in all subjects (q=0.0838) and non-obese subjects (q=0.0838), but it did not in the obese subjects (q=1.00).

    [0147] In order to elucidate whether these remarkable microbiome changes in the non-obese subjects are related to the host gene effect, the association between bacteria and genetic mutations of PNPL3, TM6SF2, MBOAT7-TMC4, and SREBF-2 using MaAsLin was analyzed. However, significant association of the four genetic mutations with three bacteria was not observed. Only Actinomyces enriched the minor allele of TM6SF2 (C/T) (q=0.169) in the non-obese subjects (FIG. 1l).

    [0148] In addition to the three variables of age, gender and BMI, the presence of type 2 diabetes mellitus (DM) was well known to affect the general changes in the microbiome (Qin J, Li Y, Cai Z, Li S, Zhu J, Zhang F, et al. A metagenome-wide association study of gut microbiota in type 2 diabetes. Nature 2012; 490:55-60.). After additional adjustment for DM, it was found that Enterobacteriaceae (p=0.00197, q=0.0616) and Faecalibacterium (p=0.00242, q=0.0707) were related to the presence of DM in all the subjects (FIG. 12a to 12d). In the non-obese subjects, not only depletion of Lachnospira (p=5.26×10.sup.−4, q=0.0676), but also the proliferation of Klebsiella (p=0.00339, q=0.141) belonging to Enterobacteriaceae in the obese subjects were also observed in the DM subjects.

    [0149] In order to understand the interaction between the microbial components and gut microbiota network characteristics in the obese and non-obese subjects, co-expression of the taxa related to the fibrosis severity was measured, and the relative abundance was shown (FIG. 2i to 2k).

    [0150] As a result, in the non-obese subjects, Veillonellaceae and Enterobacteriaceae had an inverse correlation with Ruminococcaceae (rho=−0.275 and −0.333, respectively), and Prevotellaceae showed an inverse correlation with Bacteroidaceae (rho=−0.391). However, the strong interaction between Veillonellaceae/Enterobacteriaceae and Ruminococcaceae was not observed in the obese and all subjects. In particular, the correlation of Veillonellaceae and fibrosis severity was not significant in the obese subjects and all subjects, and this suggests its specific role in progression of fibrosis in the non-obese subjects.

    [0151] In summary, the proliferation of the specific taxa according to the fibrosis severity was more pronounced in the non-obese group than in the obese group.

    [0152] 4) Observation of Fecal Metabolite Level According to Fibrosis Severity of Non-Obese and Obese NAFLD Subjects

    [0153] The non-obese and obese NAFLD subjects had different fecal metabolite levels according to the fibrosis severity. Specifically, fecal metabolites mainly related to the gut microbiota were evaluated.

    [0154] The composition of the total bile acid pool between the non-obese and obese subjects was various, and the non-obese subjects had an increased primary bile acid level according to the increased fibrosis stage (FIG. 3a).

    [0155] The total fecal bile acid level was 3 times higher in the non-obese subjects having significant fibrosis (F2-4) than the subjects without fibrosis (F0) (FIG. 3b to FIG. 3g). In particular, the cholic acid (CA), chenodeoxycholic acid (CDCA), and ursodeoxycholic acid (UDCA) levels were increased according to the fibrosis severity increased in the non-obese subjects (FIG. 3b to FIG. 3g and FIG. 13a to 13e). The lithocholic acid (LCA) and deoxycholic acid (DCA) levels were significantly high in the obese subjects having significant fibrosis, and only lithocholic acid showed significant improvement after percentage display.

    [0156] Among three SCFA, the fecal propionate level was gradually increased as fibrosis became severe in the non-obese subjects (non-obese; p=0.0032, obese; p=0.7979), and showed a significantly positive correlation with the amount of Veillonellaceae known as propionate-producing bacteria (p=0.0155) (FIG. 3h to 3j).

    [0157] On the contrary to the bile acid profile, the correlation between the significant change of fecal SCFA and its bacterial taxa was observed only in the non-obese subjects (FIG. 4b). Ruminococcus (p=0.0189), Oscillospira (p=1.57×10.sup.−4) and Desulfovibrio (p=9.76×10.sup.−4) well-known as SCFA-producing bacteria showed an inverse correlation with the fecal propionate level. The change in the fecal butyrate level according to the fibrosis severity was not found in all the non-obese and obese subjects, and the reduction of Ruminococcaceae did not affect the fecal butyrate level (non-obese, p=0.597; obese, p=0.109).

    [0158] 5) Observation of Bacterial Taxa-Metabolite Network Pattern in Non-Obese and Obese NAFLD

    [0159] A bacterial taxa-metabolite network showed a unique pattern in the non-obese and obese NAFLD. Specifically, when comparing the gut microbiota elements according to the fibrosis severity and obesity status, a clear change in the microbiome was observed only in the non-obese subjects. To investigate its core cause, NAFLD-associated genetic variant and intestinal metabolite analysis was performed.

    [0160] Based on the result, co-expression of the taxa and metabolites was evaluated, and the interaction network was shown in FIG. 4. Strong interaction between bile acids was observed in all the non-obese and obese subjects. However, the bacterial taxa and metabolite co-expression pattern according to the fibrosis severity was different in the non-obese subjects and obese subjects: The non-obese subjects showed a more significant co-expression pattern than the obese subjects.

    [0161] Interestingly, primary bile acid had an inverse correlation with Ruminococcaceae and Rikenellaceae known as indexes of healthy intestine in all the non-obese and obese subjects. Veillonellaceae exhibited a positive correlation with propionate, as well as primary bile acid. Bile acid usually has the potential to regulate growth of susceptible bacteria or to propagate relatively resistant bacteria regardless of obesity status. Nevertheless, the correlation of the intestinal bacterial taxa and fecal metabolites with the sever fibrosis was more remarkable in the non-obese NAFLD subjects than the obese subjects.

    [0162] 6) NAFLD Prediction of Non-Obese Subjects by Microbiota and Metabolite Combination

    [0163] The microbiota-metabolite combination accurately predicted significant fibrosis in the non-obese NAFLD subjects. Specifically, in order to evaluate the usefulness as a fibrosis-predicting biomarker of the gut microbiota and related fecal metabolites, AUROC for predicting significant fibrosis was compared (FIG. 5a and FIG. 5b).

    [0164] Enterobacteriaceae, Veillonellaceae, and Ruminococcaceae were selected as most representative, and significant fibrosis-related bacterial taxa. As shown in FIG. 5a, the combined bacterial marker to predict significant fibrosis yielded an AUROC of 0.824 in the non-obese subjects (0.661 for all subjects; 0.584 for obese subjects).

    [0165] In addition, Megamonas belonging to Veillonellaceae family and Ruminococcus belonging to Ruminococcaceae were selected. As shown in FIG. 5b, AUROC to predict significant fibrosis was yielded as 0.718 (0.673 for all subjects; 0.648 for obese subjects).

    [0166] As fibrosis-related metabolites, four fecal metabolites (cholic acid, chenodeoxycholic acid, ursodeoxycholic acid, and propionate) were selected, and the combination of the four metabolites predicted significant fibrosis as AUROC of 0.758 in the non-obese subjects (0.505 for all subjects; 0.520 for obese subjects).

    [0167] In case of addition of the intestinal metabolites to the bacterial marker at a family level, as shown in FIG. 5a, the predicting ability was significantly enhanced as improved AUROC of 0.977 (0.786 for all subjects; 0.609 for obese subjects). In addition, in case of addition of the intestinal metabolites to the bacterial marker at a genus level, as shown in FIG. 5b, it was enhanced as improved AUROC of 0.955 (0.590 for all subjects; 0.636 for obese subjects). The predictive ability of the novel microbiota-metabolite biomarker was significantly higher than FIB-4 widely used as a non-invasive biomarker of NAFLD.

    [0168] The result demonstrated that the diagnosis accuracy of the combination of the identified intestinal bacterial taxa and fecal metabolite, for predicting significant fibrosis in NAFLD subjects was significantly higher in the non-obese subjects than the obese subjects, and clear differences of specific bacterial taxa and large intestine metabolite between the obese and non-obese NAFLD groups could be confirmed. This result emphasizes not only the importance of the gut microbiome as a risk factor explaining the pathogenesis of non-obese NAFLD, but also the importance of application in diagnosis of the novel microbiome-metabolite combination as a non-invasive biomarker for significant fibrosis in non-obese NAFLD.